CN110739691B - Power grid branch fault prediction method and device - Google Patents
Power grid branch fault prediction method and device Download PDFInfo
- Publication number
- CN110739691B CN110739691B CN201911075832.9A CN201911075832A CN110739691B CN 110739691 B CN110739691 B CN 110739691B CN 201911075832 A CN201911075832 A CN 201911075832A CN 110739691 B CN110739691 B CN 110739691B
- Authority
- CN
- China
- Prior art keywords
- power grid
- branch
- data
- annual average
- fault
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 56
- 239000000126 substance Substances 0.000 claims description 5
- 238000009826 distribution Methods 0.000 abstract description 46
- 230000008859 change Effects 0.000 abstract description 14
- 238000004364 calculation method Methods 0.000 description 22
- 230000008569 process Effects 0.000 description 11
- 230000005540 biological transmission Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 239000011159 matrix material Substances 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 238000004146 energy storage Methods 0.000 description 3
- 150000001875 compounds Chemical class 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000010438 heat treatment Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 239000011810 insulating material Substances 0.000 description 2
- 230000009545 invasion Effects 0.000 description 2
- 238000003064 k means clustering Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 239000004576 sand Substances 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000012417 linear regression Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention discloses a power grid branch fault prediction method and a device, which can obtain at least one power grid load data set based on historical operation data of a target power grid, wherein the historical operation data is operation data of the target power grid when a first device is not additionally arranged, the obtained power grid load data sets are respectively determined as a power grid operation scene, a fault prediction formula of the target power grid is determined based on the historical operation data, operation data of each branch related to a fault rate after the first device is additionally arranged on the target power grid is obtained in each power grid operation scene, the obtained operation data of each branch is respectively input into the fault prediction formula, the fault rate of each branch after the first device is additionally arranged on the target power grid is calculated and obtained, and the change of the fault rate of each branch of a power distribution network after power distribution equipment is additionally arranged can be reflected, and a scientific and complete planning scene can be provided for the power distribution network planning work.
Description
Technical Field
The invention relates to the technical field of power distribution networks, in particular to a method and a device for predicting branch faults of a power grid.
Background
With the improvement of power transmission and distribution technology of a power grid, novel power distribution equipment such as a distributed power supply and energy storage equipment is applied to the power grid on a large scale.
The application of novel distribution equipment can effectively improve the flexibility and the economic nature of electric wire netting power transmission and distribution, simultaneously, also can show the influence to the fault rate production of each distribution lines in the electric wire netting, and technical staff need each distribution lines fault rate behind newly-increased distribution equipment in the electric wire netting carry out the analysis.
However, the prior art generally analyzes the fault rate of each distribution line in the power grid based on historical statistical data, and is not suitable for analyzing the fault rate of each distribution line after a new distribution device is added.
Disclosure of Invention
In view of the above problems, the present invention provides a method and an apparatus for predicting a branch fault of a power grid, which overcome or at least partially solve the above problems, and the technical solution is as follows:
a method for predicting a fault of a branch of a power grid comprises the following steps:
obtaining at least one power grid load data set based on historical operation data of a target power grid, wherein the historical operation data is operation data of the target power grid when a first device is not additionally arranged;
respectively determining the obtained power grid load data sets as a power grid operation scene;
determining a fault prediction formula of the target power grid based on the historical operating data;
acquiring operation data of each branch related to the fault rate after the target power grid is additionally provided with the first equipment in each power grid operation scene;
and respectively inputting the obtained operation data of each branch into the fault prediction formula, and calculating to obtain the fault rate of each branch after the target power grid is additionally provided with the first equipment.
Optionally, the obtaining at least one power grid load data set based on historical operating data of the target power grid, where the historical operating data is operating data of the target power grid when the first device is not added, includes:
obtaining power grid load data from historical operation data of a target power grid, wherein the historical operation data is operation data of the target power grid when a first device is not additionally arranged;
and clustering the obtained power grid load data to obtain at least one power grid load data set.
Optionally, determining a fault prediction formula of the target grid based on the historical operating data includes:
acquiring the annual average fault rate, the number of times of overlimit of annual average current load rate and the number of times and length of overlimit of annual average voltage of a terminal node of each branch in the target power grid from the historical operation data;
and determining a fault prediction formula of the target power grid according to the obtained annual average fault rate of each branch, the number of times of overlimit of annual average current load rate, the number of times of overlimit of annual average voltage of the terminal node and the length of the annual average voltage of the terminal node.
Optionally, the determining a fault prediction formula of the target power grid according to the obtained annual average fault rate of each branch, the obtained number of times of exceeding the annual average current load rate of each branch, and the obtained number of times of exceeding the annual average voltage of the end node and the obtained length of the annual average voltage of each end node includes:
inputting the obtained annual average fault rate, the number of times of overlimit of annual average current load rate and the number of times of overlimit of annual average voltage of the tail end node into a fault prediction formula to be determined
In (a) to obtain0、a1、a2、a3And the value of epsilon to determine a fault prediction formula;
wherein i and j are two end nodes of a branch; lambda [ alpha ]ijThe annual average failure rate of the branch ij determined by the nodes i and j;
the number of times of the out-of-limit of the annual average current load rate of the branch circuit ij is shown;
the number of times of annual average voltage overlimit of the tail end node of the branch ij is set; l isijRemoving the dimension index for the length of the branch ij; a is0Is a constant term; a is1、a2And a3Are regression coefficients respectively; ε is the error term.
Optionally, the obtaining of the operation data of each branch related to the fault rate after the target power grid is added with the first device in each power grid operation scene includes:
calculating the current load rate R of each branch circuit in the operating scene of each power grid after the first equipment is additionally arranged in the target power grids，h，ij；
Each current load factor R is calculateds，h，ijInput formula
Obtaining out-of-limit auxiliary variables of current load rates of all branches at h moment under power grid operation scene s after the first equipment is additionally arranged on the target power grid
Wherein R isfThe branch current load rate out-of-limit index is obtained;
determining probability value P corresponding to each power grid operation scenes；
Obtaining each current load rate out-of-limit auxiliary variable
And the determined probability values PsInput formula
In the middle, the number of times of the annual average current load rate out-of-limit of each branch is obtained
Optionally, the obtaining of the operation data of each branch related to the fault rate after the target power grid is added with the first device in each power grid operation scene includes:
calculating the voltage amplitude V of the tail end node of each branch in the operating scene of each power grid after the first equipment is additionally arranged in the target power grids，h，ij；
Voltage amplitude of each end nodeValue Vs，h，ijInput formula
Obtaining the out-of-limit auxiliary variable of the terminal node voltage of each branch at h moment under the power grid operation scene s after the first equipment is additionally arranged on the target power grid
Wherein the content of the first and second substances,
is an upper limit index of the node voltage out-of-limit, fVthe lower limit index is the node voltage out-of-limit;
determining probability value P corresponding to each power grid operation scenes；
Obtaining each end node voltage out-of-limit auxiliary variable
And the determined probability values PsInput formula
In the method, the number of times of the annual average terminal node voltage out-of-limit of each branch is obtained
Optionally, the obtaining of the operation data of each branch related to the fault rate after the target power grid is added with the first device in each power grid operation scene includes:
determining the length l of each branchij；
Inputting the determined branch lengths into a formula
In the method, the dimensionless index L of the length of each branch is obtainedijWherein l ismaxIs the maximum value of the branch lengths,/minIs the minimum value among the branch lengths.
A power grid branch fault prediction device comprises a first data obtaining unit, a scene determining unit, a first formula determining unit, a second data obtaining unit and a fault rate obtaining unit, wherein:
the first data obtaining unit is used for obtaining at least one power grid load data set based on historical operation data of a target power grid, wherein the historical operation data is operation data of the target power grid when first equipment is not additionally arranged;
the scene determining unit is used for determining the obtained power grid load data sets as a power grid operation scene respectively;
the first formula determination unit is used for determining a fault prediction formula of the target power grid based on the historical operation data;
the second data obtaining unit is configured to obtain, in each power grid operation scene, operation data of each branch related to a fault rate after the target power grid is added with the first device;
and the fault rate obtaining unit is used for respectively inputting the obtained operation data of each branch into the fault prediction formula and calculating and obtaining the fault rate of each branch after the target power grid is additionally provided with the first equipment.
Optionally, the first data obtaining unit specifically includes a third data obtaining unit and a data set obtaining unit, where:
the third data obtaining unit is used for obtaining power grid load data from historical operation data of a target power grid, wherein the historical operation data is operation data of the target power grid when first equipment is not additionally arranged;
the data set obtaining unit is used for clustering the obtained power grid load data to obtain at least one power grid load data set.
Optionally, the first formula determining unit specifically includes: a first obtaining unit and a second formula determining unit, wherein:
the first obtaining unit is used for obtaining fault rates of all branches in the target power grid, the out-of-limit times of current load rates, the out-of-limit times of annual average voltages of end nodes and length de-dimensionalization indexes from the historical operation data;
and the second formula determination unit is used for determining a fault prediction formula of the target power grid according to the obtained fault rate of each branch, the out-of-limit times of the current load rate, the out-of-limit times of the annual average voltage of the terminal node and the branch length dimensionless index.
Optionally, the second formula determining unit is specifically configured to:
inputting the obtained annual average fault rate, the number of times of overlimit of annual average current load rate and the number of times of overlimit of annual average voltage of the tail end node into a fault prediction formula to be determined
In (a) to obtain0、a1、a2、a3And the value of epsilon to determine a fault prediction formula;
wherein i and j are two end nodes of a branch; lambda [ alpha ]ijThe annual average failure rate of the branch ij determined by the nodes i and j;
the number of times of the out-of-limit of the annual average current load rate of the branch circuit ij is shown;
the number of times of annual average voltage overlimit of the tail end node of the branch ij is set; l isijRemoving the dimension index for the length of the branch ij; a is0Is a constant term; a is1、a2And a3Are regression coefficients respectively; ε is the error term.
Optionally, the second data obtaining unit specifically includes: a first calculation unit, a first variable obtaining unit, a first probability value determining unit, and a second obtaining unit, wherein:
the first calculation unit is configured to calculate a current load rate R of each branch in each power grid operation scenario after the first device is added to the target power grids，h，ij；
The first variable obtaining unit is used for obtaining each current load rate Rs，h，ijInput formula
Obtaining out-of-limit auxiliary variables of current load rates of all branches at h moment under power grid operation scene s after the first equipment is additionally arranged on the target power grid
Wherein R isfThe branch current load rate out-of-limit index is obtained;
the first probability value determining unit is used for determining probability values P corresponding to the power grid operation sceness；
The second obtaining unit is used for obtaining each current load rate out-of-limit auxiliary variable
And the determined probability values PsInput formula
In the middle, the number of times of the annual average current load rate out-of-limit of each branch is obtained
Optionally, the second data obtaining unit specifically includes: a second calculation unit, a second variable obtaining unit, a second probability value determining unit and a third obtaining unit, wherein:
the second calculating unit is configured to calculate a voltage amplitude V of a terminal node of each branch in each power grid operation scenario after the first device is added to the target power grids，h，ij；
The second variable obtaining unit is used for obtaining the voltage amplitude V of each end nodes，h，ijInput formula
Obtaining the out-of-limit auxiliary variable of the terminal node voltage of each branch at h moment under the power grid operation scene s after the first equipment is additionally arranged on the target power grid
Wherein the content of the first and second substances,
is an upper limit index of the node voltage out-of-limit, fVthe lower limit index is the node voltage out-of-limit;
the second probability value determining unit is used for determining the probability value P corresponding to each power grid operation scenes；
The third obtaining unit is used for obtaining the out-of-limit auxiliary variable of each end node voltage
And the determined probability values PsInput formula
In the method, the number of times of the annual average terminal node voltage out-of-limit of each branch is obtained
Optionally, the second data obtaining unit specifically includes: a branch length determining unit and a fourth obtaining unit, wherein:
the branch length determining unit is used for determining the length l of each branchij；
The fourth obtaining unit is used for inputting the determined branch lengths into a formula
In the method, the dimensionless index L of the length of each branch is obtainedijWherein l ismaxIs the maximum value of the branch lengths,/minIs the minimum value among the branch lengths.
By means of the technical scheme, the power grid branch fault prediction method and the power grid branch fault prediction device can obtain at least one power grid load data set based on historical operation data of a target power grid, wherein the historical operation data is operation data of the target power grid when a first device is not additionally arranged, the obtained power grid load data sets are respectively determined as a power grid operation scene, a fault prediction formula of the target power grid is determined based on the historical operation data, operation data of branches related to fault rates after the first device is additionally arranged on the target power grid under each power grid operation scene is obtained, the obtained operation data of each branch are respectively input into the fault prediction formula, the fault rates of the branches after the first device is additionally arranged on the target power grid are obtained through calculation, and changes of the fault rates of the branches after power distribution equipment is additionally arranged on the power distribution network can be reflected, and a scientific and complete planning scene can be provided for the power distribution network planning work.
The foregoing description is only an overview of the technical solutions of the present invention, and the embodiments of the present invention are described below in order to make the technical means of the present invention more clearly understood and to make the above and other objects, features, and advantages of the present invention more clearly understandable.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to refer to like parts throughout the drawings. In the drawings:
fig. 1 shows a flow chart of a method of grid branch fault prediction;
FIG. 2 shows a schematic block diagram of a first circuitry;
FIG. 3 illustrates a graphical representation of four power grid operating scenarios generated based on historical operating data of a first circuitry;
FIG. 4 shows a flow diagram of another grid leg fault prediction method;
fig. 5 shows a schematic diagram of a power grid branch fault prediction apparatus;
fig. 6 shows a schematic structural diagram of another grid branch fault prediction device.
Detailed Description
Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
As shown in fig. 1, the present embodiment provides a method for predicting a branch fault of a power grid, which may include the following steps:
s10, obtaining at least one power grid load data set based on historical operation data of a target power grid, wherein the historical operation data is operation data of the target power grid when a first device is not additionally arranged;
optionally, the target power grid may be a partial circuit in the power distribution network or an entire circuit of the power distribution network, which is not limited in the present invention.
The historical operation data may include operation data of each branch of the target power grid in a normal operation state, such as an annual load level operation change curve, power, node voltage (a branch may be formed by connecting two end nodes), current, fault rate and length.
It should be noted that the present invention can generate a grid load data set according to the annual load level operation change curve in the historical operation data.
Each element in each grid load data set may be a load sample, for example, a first load sample is a first element, and a second load sample is a second element.
In particular, the load sample may be a row vector consisting of load levels (i.e. load values) of the target grid at each hour of a day, for example, the load sample
Is a row vector consisting of the load level of the target grid in each hour of the day m. It should be noted that the number of dimensions in each row vector is not necessarily twenty-four, and may be other values, for example, ten or seven, which is not limited in the present invention.
In particular, the number of dimensions in each load sample (row vector) may be the same, and the time instants corresponding to the dimensions should be the same, for example, the load samples
The value of the third dimension in the second load sample refers to the load level of the distribution network at the hour one of the m days, and the value of the third dimension in the other load samples also refers to the load level of the distribution network at the hour one of the day.
Specifically, the overall load samples in each grid load dataset may form a load level curve with good continuity in time. The difference between the maximum load value and the minimum load value in any one load level curve should be within a preset threshold range, and the preset threshold range can be set by a technician according to actual conditions, which is not limited by the invention.
The first device is newly added power distribution equipment of the power distribution network, can be a distributed power supply, can be related equipment of an energy storage system, and can also be electric equipment. The number of the first devices may be one or more, and the number of the first devices is not limited in the present invention.
S20, respectively determining each obtained power grid load data set as a power grid operation scene;
it should be noted that the load level of the power grid is a main factor affecting the operation state of the power grid, the operation state of the power grid is closely related to the load level, and the power grid can adjust the operation state according to different load levels.
Specifically, the typical power grid operation scene is established to represent the annual power grid operation scene, so that the calculation process can be simplified, and the calculation efficiency can be improved.
Specifically, the load level curve formed by the increment of the whole load sample in the power grid load data set along with the time (days or hours) can be used as a power grid operation scene.
Wherein the number of grid operating scenarios is consistent with the number of grid load datasets. For example, if the number of grid load data sets is four, the number of grid operating scenarios is also four.
The load level curves with good continuity are suitable for the load level change condition (smooth change) of the target power grid in the actual operation process, and are suitable for the condition that the load level change of the target power grid in a certain period of time is usually not more than a certain value.
S30, determining a fault prediction formula of the target power grid based on the historical operation data;
optionally, in the method for predicting a fault of a branch of a power grid provided in another embodiment, step S30 may specifically include:
acquiring the annual average fault rate, the number of times of overlimit of annual average current load rate and the number of times and length of overlimit of annual average voltage of a terminal node of each branch in the target power grid from the historical operation data;
and determining a fault prediction formula of the target power grid according to the obtained annual average fault rate of each branch, the number of times of overlimit of annual average current load rate, the number of times of overlimit of annual average voltage of the terminal node and the length of the annual average voltage of the terminal node.
It should be noted that, because the number of times of exceeding the annual average current load rate of a branch, the number of times of exceeding the annual average voltage of a terminal node, and the length of the annual average current load rate of the terminal node all have significant effects on the fault rate of each branch in the power distribution network, the numerical values of historical operating data such as the number of times of exceeding the annual average current load rate, the number of times of exceeding the annual average voltage of the terminal node, the length of the annual average voltage of the terminal node, and the like can be used as known items of unknown items in the solution formula when the fault prediction formula to be determined is formulated. When the current load rate of the branch circuit is high, the line loss can cause serious heating of the circuit, and the insulating material is easy to damage; when the node voltage in the branch is out of limit, the power equipment fault is easily caused by the problems of power reverse transmission and the like; the probability of line short circuit fault caused by extreme weather, bird and animal invasion and other conditions is in direct proportion to the branch length.
And the annual average fault rate of the branch is the fault frequency of the branch in one year. For example, for the failure rate of a branch, the present invention may use the annual average failure number of the branch in the past N years as the failure rate of the branch.
The current load rate of the branch is the ratio of the actual passing current value of the branch to the maximum passing current value of the branch.
In practical application, the invention sets the limit value of the annual average current load rate of the branch circuit and the limit value of the annual average voltage of the tail end node of the branch circuit. In the running process of the power distribution network, if the annual average current load rate of a certain branch or the annual average voltage of a terminal node exceeds a corresponding limit value, the annual average current load rate of the branch or the annual average voltage of the terminal node is over-limit.
Optionally, the determining the fault prediction formula of the target power grid according to the obtained annual average fault rate of each branch, the obtained number of times of exceeding the annual average current load rate of each branch, and the obtained number of times of exceeding the annual average voltage of the end node and the obtained length of the annual average voltage of each end node may include:
inputting the obtained annual average fault rate, the number of times of overlimit of annual average current load rate and the number of times of overlimit of annual average voltage of the tail end node into a fault prediction formula to be determined
In (a) to obtain0、a1、a2、a3And the value of epsilon to determine a fault prediction formula;
wherein i and j are two end nodes of a branch; lambda [ alpha ]ijThe annual average failure rate of the branch ij determined by the nodes i and j;
the number of times of the out-of-limit of the annual average current load rate of the branch circuit ij is shown;
the number of times of annual average voltage overlimit of the tail end node of the branch ij is set; l isijRemoving the dimension index for the length of the branch ij; a is0Is a constant term; a is1、a2And a3Are regression coefficients respectively; ε is the error term.
Wherein, the length of the branch ij is descaled by a dimensionization index LijThe method is a per unit length value obtained after the branch length is subjected to dimensionless processing.
Specifically, when the annual average fault rate, the number of times of exceeding the annual average current load rate, the number of times of exceeding the annual average voltage of the end node and the length corresponding to each branch are input into a fault prediction formula to be determined, the formula can be obtained:
in the formula, NijThe number of branches for the power distribution system.
Specifically, the present invention can convert the above equation into a matrix
λ＝NA+U
In the formula (I), the compound is shown in the specification,
for each branch historical annual average fault rate vector,
in order to influence the factor matrix,
in the form of a vector of regression coefficients,
is an error vector.
Specifically, the present invention may perform regression calculation on the regression coefficient vector a by using a least square method to obtain a corresponding regression coefficient.
The error vector U is also the quantity to be solved, and after the regression calculation, the term should approach zero.
S40, acquiring operation data of each branch related to the fault rate after the first equipment is added to the target power grid under each power grid operation scene;
specifically, the load flow calculation method can perform load flow calculation on the target power grid with the first equipment added, and obtain the annual average fault rate of each branch, the number of times of exceeding the annual average current load rate, the number of times of exceeding the annual average voltage of the terminal node and the length of the terminal node.
Optionally, the obtaining of the operation data of each branch related to the fault rate after the target power grid is provided with the first device in each power grid operation scenario may include:
calculating the current load rate R of each branch circuit in the operating scene of each power grid after the first equipment is additionally arranged in the target power grids，h，ij；
Each current load factor R is calculateds，h，ijInput formula
Obtaining out-of-limit auxiliary variables of current load rates of all branches at h moment under power grid operation scene s after the first equipment is additionally arranged on the target power grid
Wherein R isfThe branch current load rate out-of-limit index is obtained;
determining probability value P corresponding to each power grid operation scenes；
Obtaining each current load rate out-of-limit auxiliary variable
And the determined probability values PsInput formula
In the middle, the number of times of the annual average current load rate out-of-limit of each branch is obtained
Specifically, the invention sets the current load rate out-of-limit auxiliary variable
The method is used for judging whether the actual current load rate of the branch exceeds the current load rate out-of-limit index of the branch or not, and if so, the current load rate out-of-limit auxiliary variable is marked as 1; if not, it is recorded as 0.
The branch current load rate out-of-limit index may be set by a technician according to an actual situation, for example, the technician may set the branch current load rate out-of-limit index to two thirds, which is not limited in the present invention.
Optionally, the obtaining of the operation data of each branch related to the fault rate after the target power grid is provided with the first device in each power grid operation scenario may include:
calculating the voltage amplitude V of the tail end node of each branch in the operating scene of each power grid after the first equipment is additionally arranged in the target power grids，h，ij；
The voltage amplitude V of each end node is measureds，h，ijInput formula
Obtaining the out-of-limit auxiliary variable of the terminal node voltage of each branch at h moment under the power grid operation scene s after the first equipment is additionally arranged on the target power grid
Wherein the content of the first and second substances,
is an upper limit index of the node voltage out-of-limit, fVthe lower limit index is the node voltage out-of-limit;
determining probability value P corresponding to each power grid operation scenes；
Obtaining each end node voltage out-of-limit auxiliary variable
And the determined probability values PsInput formula
In the method, the number of times of the annual average terminal node voltage out-of-limit of each branch is obtained
And the probability value corresponding to each power grid operation scene is the probability of the power grid operation scene appearing in all operation scenes. In particular, the probability value PsThe number of times of occurrence of the operation scene s and the number of times of occurrence of all the operation scenes.
Specifically, the invention sets an end node voltage out-of-limit auxiliary variable to judge whether the branch node voltage exceeds a reasonable operation range, if the end node voltage is higher than an upper limit or lower than a lower limit, the auxiliary variable is marked as 1, and if the end node voltage is not higher than the lower limit, the auxiliary variable is marked as 0.
The voltage out-of-limit upper and lower limit indexes can be set by a technician according to actual conditions, for example, the technician can set the voltage out-of-limit upper limit index to 1.02 and set the voltage out-of-limit lower limit index to 0.98, which is not limited in the present invention.
Optionally, the obtaining of the operation data of each branch related to the fault rate after the target power grid is provided with the first device in each power grid operation scenario may include:
determining the length l of each branchij；
Inputting the determined branch lengths into a formula
In the method, the dimensionless index L of the length of each branch is obtainedijWherein l ismaxIs the maximum value of the branch lengths,/minIs the minimum value among the branch lengths.
The method comprises the steps of obtaining a length per unit value after the branch length is subjected to dimension removal processing, namely a dimension removal index of the branch length.
And S50, respectively inputting the obtained operation data of each branch into the fault prediction formula, and calculating to obtain the fault rate of each branch after the target power grid is additionally provided with the first equipment.
Specifically, the obtained operation data of each branch circuit, namely the annual average fault rate of each branch circuit, the number of times of exceeding the annual average current load rate of each branch circuit, the number of times of exceeding the annual average voltage of the terminal node and the length of the annual average voltage of each branch circuit, can be respectively input into the determined fault prediction formula so as to obtain the annual average fault rate of each branch circuit.
Specifically, the invention can solve the failure rate by adopting a CPLEX algorithm package through a YALMIP programmed calculation model in a Matlab environment. For example, for the first circuit system of the power distribution network shown in fig. 2 (the first circuit system is a circuit system when the first device is not added to the target power grid):
the invention can input the parameter information of the first circuit system into the calculation model in advance: the active power, the reactive power and the impedance value of each node, the parameters of each branch circuit and the network topology connection relation (which can comprise the position of each node), the operating voltage level of the first circuit system, the current limit value of each branch circuit, the position and the capacity of newly added distribution equipment, the historical annual average fault frequency of each branch circuit, the annual load level change curve, the reference voltage of the first circuit system and the initial value of the reference power.
The first circuit system operation voltage level is the upper limit value and the lower limit value of the voltage of the first circuit system in safe operation.
When the types of the branches in the first circuit system are the same, the unified branch current limiting value can be input aiming at the branches with the same type; when the branches in the first circuit system are different, the invention needs to input different branch current limit values according to the branches of different models. It should be noted that each branch in the first circuit system is a line with the same type, and the present invention can input a uniform branch current limit value, such as three hundred amperes.
In this case, for the annual load level variation curve, the load level of each hour in the year may be used as a sample, that is, the annual load level variation curve may be generated according to 8760 load value sequences corresponding to time.
In the invention, the voltage (10kV) and the power value (1MVA) at the node 1 (transformer) in the first circuit system can be selected as the reference voltage and the reference power initial value of the first circuit system, respectively, that is, the voltage per unit value and the power per unit value at the node 1 are both 1.
Specifically, the detailed parameters of each node in the first circuit system can be seen in table 1 and table 2.
TABLE 1 Power at each node
TABLE 2 parameters of each branch
Among them, five tie switches (TS, tie switch) shown in fig. 2: TS1, TS2, TS3, TS4 and TS5 are provided in the branches numbered 33, 34, 35, 36 and 37, respectively.
Specifically, the historical annual average failure frequency of each branch of the first circuit system can be seen in table 3.
TABLE 3 historical annual average failure times of each branch
Specifically, the present invention executes steps S10 and S20 based on the historical operating data of the first circuit system, and may generate four grid operating scenarios as shown in fig. 3.
Specifically, the four power grid operation scenarios shown in fig. 3 may be formed by clustering power grid operation data (load data) in one year according to the present invention, and each operation scenario may be regarded as a power grid load level curve in one day (24 hours).
Specifically, the regression coefficient in the fault prediction formula determined after step S30 is executed based on the historical operating data of the first circuit system, the parameters of each node, and the parameters of each branch circuit in the present invention can be seen in table 4.
TABLE 4 multiple linear regression model regression coefficients for branch failure rates of power distribution network
Specifically, the present invention may add an intelligent soft switch to replace the tie switch TS3 in the first system, where the intelligent soft switch has a capacity of 0.5MVA and a loss factor of 0.199.
Specifically, based on the operation data of each branch after the intelligent soft switch is added to the first circuit system, the target power grid expected fault set (including the fault rate of each branch) obtained after step S50 is executed can be shown in table 5.
TABLE 5 target grid forecast failure set
It should be further noted that, because the fault rate of each branch in the expected extreme fault condition is calculated based on the operation data of the target power grid in the normal operation state, the invention establishes the link between the normal operation state and the extreme fault condition of the power distribution network, which can provide a scientific and complete planning scene (such as the fault condition) for the planning work of the power distribution network.
The method for predicting the fault of the branch of the power grid provided by this embodiment may obtain at least one power grid load dataset based on historical operation data of a target power grid, where the historical operation data is operation data of the target power grid when a first device is not added to the target power grid, respectively determine each obtained power grid load dataset as a power grid operation scene, determine a fault prediction formula of the target power grid based on the historical operation data, obtain operation data of each branch related to a fault rate after the target power grid is added with the first device in each power grid operation scene, respectively input the obtained operation data of each branch into the fault prediction formula, calculate a fault rate of each branch after the target power grid is added with the first device, and may reflect a change in the fault rate of each branch after a power distribution network is added with a power distribution device, and a scientific and complete planning scene can be provided for the power distribution network planning work.
Based on the steps shown in fig. 1, the present embodiment further provides another grid branch fault prediction method, as shown in fig. 4, the obtaining at least one grid load data set based on the historical operating data of the target grid in step S10 may include:
step S11, obtaining power grid load data from historical operation data of a target power grid, wherein the historical operation data is operation data of the target power grid when a first device is not additionally arranged;
wherein, the power grid load data is the target power grid in the normal operation state
The grid load data obtained by the present invention may include a set of grid load data sets obtained in step S10, i.e. elements in the set include a sum of elements in the grid load data sets, for example, a first grid load data set and a second grid load data set are obtained in step S10, wherein the first grid load data set is obtained
Second grid load dataset
If the grid load data is a set of the first and second grid load data sets, the grid load data may be
It should be noted that each grid load data set is obtained by clustering the grid load data according to the present invention.
And step S12, clustering the obtained power grid load data to obtain at least one power grid load data set.
Specifically, in order to select a good and representative clustering center, the euclidean distance between elements (samples) in the power grid load data can be calculated in advance:
in the formula (d)mnThe Euclidean distance between a sample m and a sample n in the power grid load data is obtained;
load level of the load sample m at the k hour; n is the dimension of each vector.
Then, the method can calculate the truncation distance d of each sample in the power grid load datacutLocal density index ρmAnd relative distance index deltam：
First, the present invention can obtain each dmnAscending according to the numerical value, and obtaining a new distance set D '═ D'1，d′2，…，d′M(M-1)D 'therein'1≤d′2…≤d′M(M-1). Get q ═ 0.02M (M-1)]([]To round the symbol), let the truncation distance dcut＝d′qThen the local density index ρmThe expression of (a) is:
in the formula, ρmIs an indicator of the local density of sample m.
Secondly, the invention can convert rhomPerforming descending arrangement according to the value size to obtain a new density set
Wherein
Then q1，q2，…，qMDenotes the set { ρ1，ρ2，…，ρMA descending subscript order of. Relative distance index deltamThe expression of (a) is:
in the formula, deltamIs a sample qmRelative distance index of (2).
Then, calculating a decision index zeta of each sample in the power grid load datam：
ζm＝ρm·δm
In the formula, ζmA decision index of the sample m is obtained;
then, the invention can determine the decision index ζmSorting in descending order to select zetamThe number of samples with larger value is the clustering number S, and ζ is calculatedmThe sample with larger value is selected as the clustering center Cs；
Thereafter, the present invention may be CsAnd (4) clustering the power grid load data by adopting a k-means clustering method as a clustering center to obtain S typical power grid operation scenes with day as a time scale and probability values corresponding to the scenes.
According to the power grid branch fault prediction method provided by the embodiment, through Euclidean distance and local density index rhomRelative distance index deltamDistance d of truncationcutAnd a decision index ζmThe clustering center of the clustering number is determined, a good and representative clustering center can be provided for the subsequent clustering process, a typical power grid operation scene is generated, and the relation between the normal operation state and the extreme fault working condition of the power distribution network is established.
Corresponding to the steps shown in fig. 1, the present embodiment proposes a power grid branch fault prediction apparatus, as shown in fig. 5, the apparatus may include a first data obtaining unit 100, a scenario determining unit 200, a first formula determining unit 300, a second data obtaining unit 400, and a fault rate obtaining unit 500, where:
the first data obtaining unit 100 is configured to obtain at least one power grid load data set based on historical operation data of a target power grid, where the historical operation data is operation data of the target power grid when a first device is not added to the target power grid;
optionally, the target power grid may be a partial circuit in the power distribution network or an entire circuit of the power distribution network, which is not limited in the present invention.
The historical operation data may include operation data of each branch of the target power grid in a normal operation state, such as an annual load level operation change curve, power, node voltage (a branch may be formed by connecting two end nodes), current, fault rate and length.
It should be noted that the present invention can generate a grid load data set according to the annual load level operation change curve in the historical operation data.
Each element in each grid load data set may be a load sample, for example, a first load sample is a first element, and a second load sample is a second element.
In particular, the load sample may be a row vector consisting of load levels (i.e., load values) of the target grid at each hour of a day. It should be noted that the number of dimensions in each row vector is not necessarily twenty-four, and may be other values, for example, ten or seven, which is not limited in the present invention.
Specifically, the number of dimensions in each load sample (row vector) may be the same, and the corresponding time instants of the dimensions should be the same.
Specifically, the overall load samples in each grid load dataset may form a load level curve with good continuity in time. The difference between the maximum load value and the minimum load value in any one load level curve should be within a preset threshold range, and the preset threshold range can be set by a technician according to actual conditions, which is not limited by the invention.
The first device is newly added power distribution equipment of the power distribution network, can be a distributed power supply, can be related equipment of an energy storage system, and can also be electric equipment. The number of the first devices may be one or more, and the number of the first devices is not limited in the present invention.
The scene determining unit 200 is configured to determine each obtained power grid load data set as a power grid operation scene;
it should be noted that the load level of the power grid is a main factor affecting the operation state of the power grid, the operation state of the power grid is closely related to the load level, and the power grid can adjust the operation state according to different load levels.
Specifically, the typical power grid operation scene is established to represent the annual power grid operation scene, so that the calculation process can be simplified, and the calculation efficiency can be improved.
Specifically, the load level curve formed by the increment of the whole load sample in the power grid load data set along with the time (days or hours) can be used as a power grid operation scene.
Wherein the number of grid operating scenarios is consistent with the number of grid load datasets. For example, if the number of grid load data sets is four, the number of grid operating scenarios is also four.
The load level curves with good continuity are suitable for the load level change condition (smooth change) of the target power grid in the actual operation process, and are suitable for the condition that the load level change of the target power grid in a certain period of time is usually not more than a certain value.
The first formula determination unit 300 is configured to determine a fault prediction formula of the target power grid based on the historical operating data;
optionally, in the device for predicting a branch fault of a power grid according to another embodiment, the first formula determining unit 300 may specifically include: a first obtaining unit and a second formula determining unit, wherein:
the first obtaining unit is used for obtaining fault rates of all branches in the target power grid, the out-of-limit times of current load rates, the out-of-limit times of annual average voltages of end nodes and length de-dimensionalization indexes from the historical operation data;
and the second formula determination unit is used for determining a fault prediction formula of the target power grid according to the obtained fault rate of each branch, the out-of-limit times of the current load rate, the out-of-limit times of the annual average voltage of the terminal node and the branch length dimensionless index.
It should be noted that, because the number of times of exceeding the annual average current load rate of a branch, the number of times of exceeding the annual average voltage of a terminal node, and the length of the annual average current load rate of the terminal node all have significant effects on the fault rate of each branch in the power distribution network, the numerical values of historical operating data such as the number of times of exceeding the annual average current load rate, the number of times of exceeding the annual average voltage of the terminal node, the length of the annual average voltage of the terminal node, and the like can be used as known items of unknown items in the solution formula when the fault prediction formula to be determined is formulated. When the current load rate of the branch circuit is high, the line loss can cause serious heating of the circuit, and the insulating material is easy to damage; when the node voltage in the branch is out of limit, the power equipment fault is easily caused by the problems of power reverse transmission and the like; the probability of line short circuit fault caused by extreme weather, bird and animal invasion and other conditions is in direct proportion to the branch length.
And the annual average fault rate of the branch is the fault frequency of the branch in one year. For example, for the failure rate of a branch, the present invention may use the annual average failure number of the branch in the past N years as the failure rate of the branch.
The current load rate of the branch is the ratio of the actual passing current value of the branch to the maximum passing current value of the branch.
In practical application, the invention sets the limit value of the annual average current load rate of the branch circuit and the limit value of the annual average voltage of the tail end node of the branch circuit. In the running process of the power distribution network, if the annual average current load rate of a certain branch or the annual average voltage of a terminal node exceeds a corresponding limit value, the annual average current load rate of the branch or the annual average voltage of the terminal node is over-limit.
Optionally, the second formula determining unit may be specifically configured to:
inputting the obtained annual average fault rate, the number of times of overlimit of annual average current load rate and the number of times of overlimit of annual average voltage of the tail end node into a fault prediction formula to be determined
In (a) to obtain0、a1、a2、a3And the value of epsilon to determine a fault prediction formula;
wherein i and j are two end nodes of a branch; lambda [ alpha ]ijThe annual average failure rate of the branch ij determined by the nodes i and j;
the number of times of the out-of-limit of the annual average current load rate of the branch circuit ij is shown;
the number of times of annual average voltage overlimit of the tail end node of the branch ij is set; l isijRemoving the dimension index for the length of the branch ij; a is0Is a constant term; a is1、a2And a3Are regression coefficients respectively; ε is the error term.
Wherein, the length of the branch ij is descaled by a dimensionization index LijThe method is a per unit length value obtained after the branch length is subjected to dimensionless processing.
Specifically, when the annual average fault rate, the number of times of exceeding the annual average current load rate, the number of times of exceeding the annual average voltage of the end node and the length corresponding to each branch are input into a fault prediction formula to be determined, the formula can be obtained:
in the formula, NijThe number of branches for the power distribution system.
Specifically, the present invention can convert the above equation into a matrix
λ＝NA+U
In the formula (I), the compound is shown in the specification,
for each branch historical annual average fault rate vector,
in order to influence the factor matrix,
in the form of a vector of regression coefficients,
is an error vector.
Specifically, the present invention may perform regression calculation on the regression coefficient vector a by using a least square method to obtain a corresponding regression coefficient.
The error vector U is also the quantity to be solved, and after the regression calculation, the term should approach zero.
The second data obtaining unit 400 is configured to obtain, in each power grid operation scenario, operation data of each branch related to a fault rate after the target power grid is added with the first device;
specifically, the load flow calculation method can perform load flow calculation on the target power grid with the first equipment added, and obtain the annual average fault rate of each branch, the number of times of exceeding the annual average current load rate, the number of times of exceeding the annual average voltage of the terminal node and the length of the terminal node.
Optionally, the second data obtaining unit 400 may specifically include: a first calculation unit, a first variable obtaining unit, a first probability value determining unit, and a second obtaining unit, wherein:
the first calculation unit is configured to calculate a current load rate R of each branch in each power grid operation scenario after the first device is added to the target power grids，h，ij
The first variable obtaining unit is used for obtaining each current load rate Rs，h，ijInput formula
Obtaining out-of-limit auxiliary variables of current load rates of all branches at h moment under power grid operation scene s after the first equipment is additionally arranged on the target power grid
Wherein R isfThe branch current load rate out-of-limit index is obtained;
the first probability value determining unit is used for determining probability values P corresponding to the power grid operation sceness；
The second obtaining unit is used for obtaining each current load rate out-of-limit auxiliary variable
And the determined probability values PsInput formula
In the middle, the number of times of the annual average current load rate out-of-limit of each branch is obtained
And the probability value corresponding to each power grid operation scene is the probability of the power grid operation scene appearing in all operation scenes. In particular, the probability value PsThe number of times of occurrence of the operation scene s and the number of times of occurrence of all the operation scenes.
Specifically, the invention sets the current load rate out-of-limit auxiliary variable
For determining the actual current load rate of the branchWhether the current load rate of the branch is exceeded or not is judged, if yes, the current load rate out-of-limit auxiliary variable is marked as 1; if not, it is recorded as 0.
The branch current load rate out-of-limit index can be formulated by technicians according to actual conditions.
Optionally, the second data obtaining unit 400 may specifically include: a second calculation unit, a second variable obtaining unit, a second probability value determining unit and a third obtaining unit, wherein:
the second calculating unit is configured to calculate a voltage amplitude V of a terminal node of each branch in each power grid operation scenario after the first device is added to the target power grids，h，ij；
The second variable obtaining unit is used for obtaining the voltage amplitude V of each end nodes，h，ijInput formula
Obtaining the out-of-limit auxiliary variable of the terminal node voltage of each branch at h moment under the power grid operation scene s after the first equipment is additionally arranged on the target power grid
Wherein the content of the first and second substances,
is an upper limit index of the node voltage out-of-limit, fVthe lower limit index is the node voltage out-of-limit;
the second probability value determining unit is used for determining the probability value P corresponding to each power grid operation scenes；
The third obtaining unit is used for obtaining the out-of-limit auxiliary variable of each end node voltage
And the determined probability values PsInput formula
In the method, the number of times of the annual average terminal node voltage out-of-limit of each branch is obtained
Specifically, the invention sets an end node voltage out-of-limit auxiliary variable to judge whether the branch node voltage exceeds a reasonable operation range, if the end node voltage is higher than an upper limit or lower than a lower limit, the auxiliary variable is marked as 1, and if the end node voltage is not higher than the lower limit, the auxiliary variable is marked as 0.
The voltage out-of-limit upper and lower limit indexes can be formulated by technicians according to actual conditions, and the invention is not limited to this.
Optionally, the second data obtaining unit 400 may specifically include: a branch length determining unit and a fourth obtaining unit, wherein:
the branch length determining unit is used for determining the length l of each branchij；
The fourth obtaining unit is used for inputting the determined branch lengths into a formula
In the method, the dimensionless index L of the length of each branch is obtainedijWherein l ismaxIs the maximum value of the branch lengths,/minIs the minimum value among the branch lengths.
The method comprises the steps of obtaining a length per unit value after the branch length is subjected to dimension removal processing, namely a dimension removal index of the branch length.
The failure rate obtaining unit 500 is configured to input the obtained operation data of each branch into the failure prediction formula, and calculate and obtain the failure rate of each branch after the target power grid is added with the first device.
Specifically, the obtained operation data of each branch circuit, namely the annual average fault rate of each branch circuit, the number of times of exceeding the annual average current load rate of each branch circuit, the number of times of exceeding the annual average voltage of the terminal node and the length of the annual average voltage of each branch circuit, can be respectively input into the determined fault prediction formula so as to obtain the annual average fault rate of each branch circuit.
The power grid branch fault prediction apparatus provided in this embodiment may obtain at least one power grid load dataset based on historical operation data of a target power grid, where the historical operation data is operation data of the target power grid when a first device is not added to the target power grid, respectively determine each obtained power grid load dataset as a power grid operation scene, determine a fault prediction formula of the target power grid based on the historical operation data, obtain operation data of each branch related to a fault rate after the target power grid is added to the first device in each power grid operation scene, respectively input the obtained operation data of each branch to the fault prediction formula, calculate a fault rate of each branch after the target power grid is added to the first device, and may reflect a change in the fault rate of each branch after a power distribution network is added to the power distribution device, and a scientific and complete planning scene can be provided for the power distribution network planning work.
Based on the apparatus shown in fig. 5, this embodiment further provides another apparatus for predicting a branch fault of a power grid, as shown in fig. 6, the first data obtaining unit 100 may specifically include a third data obtaining unit 110 and a data set obtaining unit 120, where:
the third data obtaining unit 110 is configured to obtain power grid load data from historical operation data of a target power grid, where the historical operation data is operation data of the target power grid when no first device is added to the target power grid;
wherein, the power grid load data is the target power grid in the normal operation state
The grid load data obtained by the present invention may include a set of each grid load data set obtained in the first data obtaining unit 100, that is, the elements in the set include the sum of the elements in each grid load data set.
It should be noted that each grid load data set is obtained by clustering the grid load data according to the present invention.
The data set obtaining unit 120 is configured to cluster the obtained grid load data to obtain at least one grid load data set.
Specifically, in order to select a good and representative clustering center, the euclidean distance between elements (samples) in the power grid load data can be calculated in advance:
in the formula (d)mnThe Euclidean distance between a sample m and a sample n in the power grid load data is obtained;
load level of the load sample m at the k hour; n is the dimension of each vector.
Then, the method can calculate the truncation distance d of each sample in the power grid load datacutLocal density index ρmAnd relative distance index deltam：
First, the present invention can obtain each dmnAscending according to the numerical value, and obtaining a new distance set D '═ D'1，d′2，…，d′M(M-1)D 'therein'1≤d′2…≤d′M(M-1). Get q ═ 0.02M (M-1)]([]To round the symbol), let the truncation distance dcut＝d′qThen the local density index ρmThe expression of (a) is:
in the formula, ρmLocal density index for sample m。
Secondly, the invention can convert rhomPerforming descending arrangement according to the value size to obtain a new density set
Wherein
Then q1，q2，…，qMDenotes the set { ρ1，ρ2，…，ρMA descending subscript order of. Relative distance index deltamThe expression of (a) is:
in the formula, deltamIs a sample qmRelative distance index of (2).
Then, calculating a decision index zeta of each sample in the power grid load datam：
ζm＝ρm·δm
In the formula, ζmA decision index of the sample m is obtained;
then, the invention can determine the decision index ζmSorting in descending order to select zetamThe number of samples with larger value is the clustering number S, and ζ is calculatedmThe sample with larger value is selected as the clustering center Cs；
Thereafter, the present invention may be CsAnd (4) clustering the power grid load data by adopting a k-means clustering method as a clustering center to obtain S typical power grid operation scenes with day as a time scale and probability values corresponding to the scenes.
The power grid branch fault prediction device provided by the embodiment adopts the Euclidean distance and the local density index rhomRelative distance index deltamDistance d of truncationcutAnd a decision index ζmThe calculation of the cluster number and the determination of the cluster center can provide good and representative cluster centers for the subsequent clustering process,and generating a typical power grid operation scene, and establishing a relation between a normal operation state and an extreme fault working condition of the power distribution network.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in the process, method, article, or apparatus that comprises the element.
The above are merely examples of the present application and are not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.
Claims (10)
1. A method for predicting a fault of a branch of a power grid, comprising:
obtaining at least one power grid load data set based on historical operation data of a target power grid, wherein the historical operation data is operation data of the target power grid when a first device is not additionally arranged;
respectively determining the obtained power grid load data sets as a power grid operation scene;
determining a fault prediction formula of the target power grid based on the historical operating data;
acquiring operation data of each branch related to the fault rate after the target power grid is additionally provided with the first equipment in each power grid operation scene;
and respectively inputting the obtained operation data of each branch into the fault prediction formula, and calculating to obtain the fault rate of each branch after the target power grid is additionally provided with the first equipment.
2. The method according to claim 1, wherein the obtaining at least one grid load dataset based on historical operating data of the target grid, the historical operating data being operating data of the target grid without the addition of the first device, comprises:
obtaining power grid load data from historical operation data of a target power grid, wherein the historical operation data is operation data of the target power grid when a first device is not additionally arranged;
and clustering the obtained power grid load data to obtain at least one power grid load data set.
3. The method of claim 1, wherein determining a fault prediction formula for the target grid based on the historical operating data comprises:
acquiring the annual average fault rate, the number of times of overlimit of annual average current load rate and the number of times and length of overlimit of annual average voltage of a terminal node of each branch in the target power grid from the historical operation data;
and determining a fault prediction formula of the target power grid according to the obtained annual average fault rate of each branch, the number of times of overlimit of annual average current load rate, the number of times of overlimit of annual average voltage of the terminal node and the length of the annual average voltage of the terminal node.
4. The method according to claim 3, wherein the determining a fault prediction formula of the target power grid according to the obtained annual average fault rate, the number of times of exceeding annual average current load rate, the number of times of exceeding annual average voltage of the end node and the length comprises:
inputting the obtained annual average fault rate, the number of times of overlimit of annual average current load rate and the number of times of overlimit of annual average voltage of the tail end node into a fault prediction formula to be determined
In (a) to obtain0、a1、a2、a3And the value of epsilon to determine a fault prediction formula;
wherein i and j are two end nodes of a branch; lambda [ alpha ]ijThe annual average failure rate of the branch ij determined by the nodes i and j;
the number of times of the out-of-limit of the annual average current load rate of the branch circuit ij is shown;
the number of times of annual average voltage overlimit of the tail end node of the branch ij is set; l isijRemoving the dimension index for the length of the branch ij; a is0Is a constant term; a is1、a2And a3Are regression coefficients respectively; ε is the error term.
5. The method according to claim 3, wherein the obtaining of the operation data of each branch related to the fault rate after the target grid is added with the first device in each grid operation scenario comprises:
calculating the current load rate R of each branch circuit in the operating scene of each power grid after the first equipment is additionally arranged in the target power grids，h，ij；
Each current load factor R is calculateds，h，ijInput formula
Obtaining out-of-limit auxiliary variables of current load rates of all branches at h moment under power grid operation scene s after the first equipment is additionally arranged on the target power grid
Wherein R isfThe branch current load rate out-of-limit index is obtained;
determining probability value P corresponding to each power grid operation scenesThe probability value PsThe probability of occurrence of each power grid operation scene in all operation scenes is given;
obtaining each current load rate out-of-limit auxiliary variable
And the determined probability values PsInput formula
In the middle, the number of times of the annual average current load rate out-of-limit of each branch is obtained
6. The method according to claim 3, wherein the obtaining of the operation data of each branch related to the fault rate after the target grid is added with the first device in each grid operation scenario comprises:
calculating the voltage amplitude V of the tail end node of each branch in the operating scene of each power grid after the first equipment is additionally arranged in the target power grids，h，ij；
The voltage amplitude V of each end node is measureds，h，ijInput formula
Obtaining the out-of-limit auxiliary variable of the terminal node voltage of each branch at h moment under the power grid operation scene s after the first equipment is additionally arranged on the target power grid
Wherein the content of the first and second substances,
is an upper limit index of the node voltage out-of-limit, fVthe lower limit index is the node voltage out-of-limit;
determining probability value P corresponding to each power grid operation scenesThe probability value PsThe probability of occurrence of each power grid operation scene in all operation scenes is given;
obtaining each end node voltage out-of-limit auxiliary variable
And the determined probability values PsInput formula
In the method, the number of times of the annual average terminal node voltage out-of-limit of each branch is obtained
7. The method according to claim 3, wherein the obtaining of the operation data of each branch related to the fault rate after the target grid is added with the first device in each grid operation scenario comprises:
determining the length l of each branchij；
Inputting the determined branch lengths into a formula
In the method, the dimensionless index L of the length of each branch is obtainedijWherein l ismaxIs the maximum value of the branch lengths,/minIs the minimum value among the branch lengths.
8. A power grid branch fault prediction device is characterized by comprising a first data obtaining unit, a scene determining unit, a first formula determining unit, a second data obtaining unit and a fault rate obtaining unit, wherein:
the first data obtaining unit is used for obtaining at least one power grid load data set based on historical operation data of a target power grid, wherein the historical operation data is operation data of the target power grid when first equipment is not additionally arranged;
the scene determining unit is used for determining the obtained power grid load data sets as a power grid operation scene respectively;
the first formula determination unit is used for determining a fault prediction formula of the target power grid based on the historical operation data;
the second data obtaining unit is configured to obtain, in each power grid operation scene, operation data of each branch related to a fault rate after the target power grid is added with the first device;
and the fault rate obtaining unit is used for respectively inputting the obtained operation data of each branch into the fault prediction formula and calculating and obtaining the fault rate of each branch after the target power grid is additionally provided with the first equipment.
9. The apparatus according to claim 8, wherein the first data obtaining unit specifically comprises a third data obtaining unit and a data set obtaining unit, wherein:
the third data obtaining unit is used for obtaining power grid load data from historical operation data of a target power grid, wherein the historical operation data is operation data of the target power grid when first equipment is not additionally arranged;
the data set obtaining unit is used for clustering the obtained power grid load data to obtain at least one power grid load data set.
10. The apparatus according to claim 8, wherein the first formula determining unit specifically includes: a first obtaining unit and a second formula determining unit, wherein:
the first obtaining unit is used for obtaining fault rates of all branches in the target power grid, the out-of-limit times of current load rates, the out-of-limit times of annual average voltages of end nodes and length de-dimensionalization indexes from the historical operation data;
and the second formula determination unit is used for determining a fault prediction formula of the target power grid according to the obtained fault rate of each branch, the out-of-limit times of the current load rate, the out-of-limit times of the annual average voltage of the terminal node and the branch length dimensionless index.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911075832.9A CN110739691B (en) | 2019-11-06 | 2019-11-06 | Power grid branch fault prediction method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911075832.9A CN110739691B (en) | 2019-11-06 | 2019-11-06 | Power grid branch fault prediction method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110739691A CN110739691A (en) | 2020-01-31 |
| CN110739691B true CN110739691B (en) | 2021-05-14 |
Family
ID=69272327
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911075832.9A Active CN110739691B (en) | 2019-11-06 | 2019-11-06 | Power grid branch fault prediction method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110739691B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104009467A (en) * | 2014-05-13 | 2014-08-27 | 南京邮电大学 | Meter and power distribution network reliability assessment and prediction method for pre-arranging power outage influence |
| CN107516170A (en) * | 2017-08-30 | 2017-12-26 | 东北大学 | A kind of difference self-healing control method based on probability of equipment failure and power networks risk |
| CN109245091A (en) * | 2018-09-27 | 2019-01-18 | 东南大学 | A kind of two_layer strategies dispatching method towards sale of electricity company |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105069535B (en) * | 2015-08-19 | 2020-07-24 | 中国电力科学研究院 | Power distribution network operation reliability prediction method based on ARIMA model |
-
2019
- 2019-11-06 CN CN201911075832.9A patent/CN110739691B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104009467A (en) * | 2014-05-13 | 2014-08-27 | 南京邮电大学 | Meter and power distribution network reliability assessment and prediction method for pre-arranging power outage influence |
| CN107516170A (en) * | 2017-08-30 | 2017-12-26 | 东北大学 | A kind of difference self-healing control method based on probability of equipment failure and power networks risk |
| CN109245091A (en) * | 2018-09-27 | 2019-01-18 | 东南大学 | A kind of two_layer strategies dispatching method towards sale of electricity company |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110739691A (en) | 2020-01-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109145957B (en) | Method and device for identifying and processing abnormal indexes of power distribution network based on big data | |
| CN108876054B (en) | Short-term power load prediction method based on improved genetic algorithm optimization extreme learning machine | |
| CN107516170B (en) | Difference self-healing control method based on equipment failure probability and power grid operation risk | |
| CN114460445B (en) | Transformer aging unavailability evaluation method considering aging threshold and service life | |
| CN109613338B (en) | Low-voltage user loop impedance estimation method based on unitary model | |
| CN104037943A (en) | Method and system for monitoring voltage and capable of improving power grid voltage quality | |
| CN109190242B (en) | Method for estimating line impedance of household section | |
| CN114252103B (en) | Fusion power station operation fault prediction method | |
| CN112132283B (en) | Genetic algorithm-based non-signal injection type household transformer topological relation identification method | |
| CN113904322A (en) | Low-voltage distribution network topology generation method based on current and voltage | |
| CN113724101B (en) | Table relation identification method and system, equipment and storage medium | |
| CN113379005B (en) | Intelligent energy management system and method for power grid power equipment | |
| CN111221811A (en) | Low-voltage distribution network line parameter estimation method based on centralized meter reading system | |
| CN110739691B (en) | Power grid branch fault prediction method and device | |
| CN107563641B (en) | Disaster-resistant multi-scene differentiation planning method for power distribution network considering disaster preference | |
| CN104732107B (en) | Using medium parameter as the residual life of sleeve of transformer Forecasting Methodology for assessing parameter | |
| CN107301499B (en) | Distribution feeder statistical line loss rate data cleaning method based on AMI data | |
| CN111864728B (en) | Important equipment identification method and system for reconfigurable power distribution network | |
| CN112485604A (en) | Power distribution network fault current calculation method | |
| CN110108983B (en) | Intelligent cable state online measurement and control management method | |
| CN112241812B (en) | Topology identification method for low-voltage distribution network based on single-side optimization and genetic algorithm cooperation | |
| CN108564249B (en) | Power distribution network confidence peak clipping benefit evaluation method considering distributed photovoltaic randomness | |
| CN115494382A (en) | Method, system, equipment and storage medium for judging superior-inferior relation of intelligent switch | |
| CN112946369B (en) | Cable aging identification method using power line communication device | |
| CN113887144A (en) | Electric power communication optical cable comprehensive availability analysis method considering service characteristics |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |